Methodology
How a vendor score is calculated
Every score in this index is computed from the ratings below — nobody edits a vendor’s total directly. If a criterion doesn’t apply to a vendor (for example, a vendor with no public discount code has nothing to score for “code verification”), it’s marked not applicable and excluded from that vendor’s average rather than counted against it.
The scale
| Strong | 2 / 2 points |
| Partial | 1.5 / 2 points |
| Unconfirmed | 1 / 2 points |
| Weak / red flag | 0 / 2 points |
A vendor’s total = the average of its applicable criterion points, scaled to a 0–10 range. Five criteria, equally weighted, unless a criterion is not applicable.
Where the lab report resolves
Do the published Certificates of Analysis resolve on a laboratory's own domain, or are they files the seller hosts?
This is the single highest-value check a non-specialist can make, and it takes about ten seconds. A certificate that exists only as a PDF on the seller's own server can be edited by that seller. One that resolves on the laboratory's portal, carrying the lab's own reference and naming the client and batch, cannot. We rate this on where the files actually live, not on what the vendor says about testing.
Depth of the published set
How many reports are published, and is the full distribution visible — including the weak results?
One certificate proves one lot. A published set shows whether results hold across products and over time, and a set that contains no weak results has been filtered — which tells you about the filter rather than the products. We count what is published and note whether low results are left visible.
Laboratory credential, verified
Has the testing laboratory's accreditation tier actually been checked, and what is it?
ISO 17025 accreditation is a materially stronger claim than a CLIA waiver, which is itself stronger than an unnamed lab. We rate what we have verified rather than what is asserted: a vendor whose lab we checked and found lightly credentialled scores above one whose lab we have not been able to check at all, because a verified weakness is worth more to a reader than an unverified strength.
Public track record, independently verifiable
How far back can this vendor's own storefront be seen in a public archive that nobody involved controls?
A seller can say it has been trading for years. A public web archive is the cheapest way to check, because the archive is run by a third party and the seller cannot edit it. We take the earliest archived capture that shows an actual storefront — a catalogue, products, a cart — not merely the domain existing, because a parked page proves nothing. Read what this measures narrowly: it is the length of the record a stranger can verify, not a judgement of quality, and a long record is not a guarantee of anything. It also has a real limitation we state rather than hide — archive coverage is uneven, so a vendor may have been trading before its first capture, or may be missing from the archive entirely while operating normally. Where that is the case we say so on the vendor's own page and give the domain registration date alongside, so a reader can weigh both.
Shipping origin / customs exposure
Does the order ship domestically within Canada, or does it cross an international border?
Domestic shipping means the parcel never meets a customs officer — no secondary-inspection risk, no cross-border hold. This is a structural fact about where the vendor ships from, stated per the vendor's own published shipping policy, not a claim about product quality.
Payment reversibility
Does the vendor accept a payment method a buyer can dispute (credit card, Interac) or is it irreversible only (crypto)?
If every payment option is irreversible, the buyer carries all the risk if something goes wrong. We mark this unconfirmed rather than guessing wherever we don't have a checked, current answer — see each vendor's page for what's actually been verified.
Discount code — last cart-tested
Is there a live, publicly-checkable discount code, and how recently was it actually tested?
Published as a dated factual record, and deliberately NOT part of the credibility score. Whether a vendor runs a discount code says nothing about whether its products are what the label says, and until 2026-10-05 this criterion was the only thing separating the top of this table — which made the score a measure of couponing rather than of credibility. The record is still here, with the same evidence, because it is useful; it just no longer moves the number.
Discount code — math checks out
Is there a current, publicly-stated discount, and does it apply at the stated rate?
A code that claims 10% and gives 5%, or quietly expired, is worse than no code. Published as a dated record for the same reason as the criterion above, and excluded from the score for the same reason.
What this rubric deliberately doesn’t do
- It doesn’t score product effects, potency claims, or anything about what a peptide does — that’s outside what a vendor-vetting checklist can verify.
- It doesn’t treat “no public discount code” as a strike against a vendor — plenty of legitimate vendors don’t run one.
- It doesn’t guess at facts we haven’t verified. Where we don’t have a checked, current answer, the criterion is marked unconfirmed rather than assumed either way.
- It isn’t designed around any particular vendor finishing first — see each vendor’s page for the evidence behind its number.